Testing scientific models using Qualitative Reasoning: Application to cellulose hydrolysis.

With the accumulation of scientific information in natural science, even experts can find difficult to keep integrating new piece of information. It is critical to explore modelling solutions able to capture information scattered in publications as a computable representation form. Traditional modelling techniques are important in that regard, but relying on numerical information comes with limitations for integrating results from distinct studies, high-level representations can be more suited. We present an approach to stepwise construct mechanistic explanation from selected scientific papers using the Qualitative Reasoning framework. As a proof of concept, we apply the approach to modelling papers about cellulose hydrolysis mechanism, focusing on the causal explanations for the decreasing of hydrolytic rate. Two explanatory QR models are built to capture classical explanations for the phenomenon. Our results show that none of them provides sufficient explanation for a set of basic experimental observations described in the literature. Combining the two explanations into a third one allowed to get a new and sufficient explanation for the experimental results. In domains where numerical data are scarce and strongly related to the experimental conditions, this approach can aid assessing the conceptual validity of an explanation and support integration of knowledge from different sources.

Contrasted enzymatic cocktails reveal the importance of cellulases and hemicellulases activity ratios for the hydrolysis of cellulose in presence of xylans.

Various enzymatic cocktails were produced from two Trichoderma reesei strains, a cellulase hyperproducer strain and a strain with ß-glucosidase activity overexpression. By using various carbon sources (lactose, glucose, xylose, hemicellulosic hydrolysate) for strains growth, contrasted enzymatic activities were obtained. The enzymatic cocktails presented various levels of efficiency for the hydrolysis of cellulose Avicel into glucose, in presence of xylans, or not. These latter were also hydrolyzed with different extents according to cocktails. The most efficient cocktails (TR1 and TR3) on Avicel were richer in filter paper activity (FPU) and presented a low ratio FPU/ß-glucosidase activity. Cocktails TR2 and TR5 which were produced on the higher amount of hemicellulosic hydrolysate, possess both high xylanase and ß-xylosidase activities, and were the most efficient for xylans hydrolysis. When hydrolysis of Avicel was conducted in presence of xylans, a decrease of glucose release occurred for all cocktails compared to hydrolysis of Avicel alone. Mixing TR1 and TR5 cocktails with two different ratios of proteins (1/1 and 1/4) resulted in a gain of efficiency for glucose release during hydrolysis of Avicel in presence of xylans compared to TR5 alone. Our results demonstrate the importance of combining hemicellulase and cellulase activities to improve the yields of glucose release from Avicel in presence of xylans. In this context, strategies involving enzymes production with carbon sources comprising mixed C5 and C6 sugars or combining different cocktails produced on C5 or on C6 sugars are of interest for processes developed in the context of lignocellulosic biorefinery.